Firmware Update

[ahuber1]

Hi,

it's really strange after i create a different firmware now i'm unable to create some working.
I already have deleted all downloaded files and follow the instructions without success.

[ahuber1]

i think the compiler is the right one.
arm-none-eabi-gcc (GNU Tools for ARM Embedded Processors) 4.9.3 20141119 (release) [ARM/embedded-4_9-branch revision 218278]

[chad]

i think the compiler is the right one.
arm-none-eabi-gcc (GNU Tools for ARM Embedded Processors) 4.9.3 20141119 (release) [ARM/embedded-4_9-branch revision 218278]

Yes, that should be the right one.
gcc version 4.9.3 20141119 (release) [ARM/embedded-4_9-branch revision 218278] (GNU Tools for ARM Embedded Processors)
https://launchpad.net/gcc-arm-embedded/ ... -q4-major/

What is the issue you are having? Steady blue light still after flashing and restarting? Have you edited the firmware source at all or is it clean (run 'git status' to be sure)?

[ahuber1]

Yes steady blue light, read light blink one time after a cold start.
The green blink two times, first time normal second time weak
I use master branch without any changes.

make all PLATFORM=CF1 CLOAD=0 PYTHON2=python

Crazyflie Nano (1.0) build!
Build 54:b4b7a369d1f4 (2015.08.1-54) CLEAN
Version extracted from git
text data bss dec hex filename
66068 356 19796 86220 150cc cf1.elf

[chad]

Yes steady blue light, read light blink one time after a cold start.
The green blink two times, first time normal second time weak

Have you edited the firmware source at all or is it clean (run 'git status' to be sure)?

[ahuber1]

On branch master
Your branch is up-to-date with 'origin/master'.
nothing to commit, working directory clean

[chad]

On branch master
Your branch is up-to-date with 'origin/master'.
nothing to commit, working directory clean

Hmmm... Curious. Your battery is OK. Still holds a good charge? Did blue-light start being static again in-between firmware builds or did you flash, and it ran OK then the next time you powered on it got stuck?

I just updated my master branch. The last commit as of this writing is:

Code: Select all

commit b4b7a369d1f45fa617644b1f3875d931eba65241
Author: Marcus Eliasson <marcus@bitcraze.se>
Date:   Sun Dec 20 19:54:31 2015 +0100

    Option for disabling auto initialization of found decks

My repository is clean (like yours) and when I run a build I end up with:

Code: Select all

Build 54:b4b7a369d1f4 (2015.08.1-54) CLEAN

The md5 for the cf1.bin file is:

Code: Select all

MD5 (cf1.bin) = 58cc943455d9e6236a0c664d29bffbd0

Is your repository state and build output (build number and md5 checksum of cf1.bin) the same as mine?

[ahuber1]

Yes, battery is OK.
When i flash it with 2005.08.01.bin everything works as expected.
The lights are the same for cold and warm boot.
The last commit is still the same.
But my HASH is still different.

Code: Select all

MD5 (cf1.bin) = 11cb7516ac0d7713816f256cc054455b

When i create the HASH in Bitcaze VM it looks like same

Code: Select all

11cb7516ac0d7713816f256cc054455b  cf1.bin

My Environment variables are

Code: Select all

declare -x Apple_PubSub_Socket_Render="/private/tmp/com.apple.launchd.RD7JlLpoCE/Render"
declare -x DISPLAY="/private/tmp/com.apple.launchd.sGd8SOjpyO/org.macosforge.xquartz:0"
declare -x HOME="/Users/ahuber"
declare -x JAVA_HOME="/Library/Java/JavaVirtualMachines/jdk1.7.0_76.jdk/Contents/Home"
declare -x LANG="de_DE.UTF-8"
declare -x LOGNAME="ahuber"
declare -x M2="/opt/maven/bin"
declare -x M2_HOME="/opt/maven"
declare -x OLDPWD="/Users/ahuber/Documents/Projects/crazyflie/crazyflie-firmware"
declare -x PATH="/Users/ahuber/Downloads/GCC-ARM/gcc-arm-none-eabi-4_9-2014q4/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin"
declare -x PREV_PATH="/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin"
declare -x PWD="/Users/ahuber/Documents/Projects/crazyflie/crazyflie-firmware"
declare -x SHELL="/bin/bash"
declare -x SHLVL="1"
declare -x SSH_AUTH_SOCK="/private/tmp/com.apple.launchd.wXWlNMtRcy/Listeners"
declare -x SVN_HOME="/Applications/Xcode.app/Contents/Developer/usr/subversion-1.6/bin"
declare -x TERM="xterm-256color"
declare -x TERM_PROGRAM="Apple_Terminal"
declare -x TERM_PROGRAM_VERSION="361.1"
declare -x TERM_SESSION_ID="1B4FB4E1-3C2C-4F7F-9330-9C5FF929F420"
declare -x TMPDIR="/var/folders/rm/yzfbkl4j7k19jxgzrxb27dbw0000gn/T/"
declare -x USER="ahuber"
declare -x XPC_FLAGS="0x0"
declare -x XPC_SERVICE_NAME="0"
declare -x __CF_USER_TEXT_ENCODING="0x1F5:0x0:0x3"

I have already minimized the PATH variable to a minimum.

[chad]

Very curious indeed! In my experience, when the firmware didn't build as I expected, it was usually because another gcc was doing the compilation. In your case though, given your $PATH, I can't see what else would be getting called. This is very strange to me!

[chad]

My Environment variables are

Code: Select all

declare -x Apple_PubSub_Socket_Render="/private/tmp/com.apple.launchd.RD7JlLpoCE/Render"
declare -x DISPLAY="/private/tmp/com.apple.launchd.sGd8SOjpyO/org.macosforge.xquartz:0"
declare -x HOME="/Users/ahuber"
declare -x JAVA_HOME="/Library/Java/JavaVirtualMachines/jdk1.7.0_76.jdk/Contents/Home"
declare -x LANG="de_DE.UTF-8"
declare -x LOGNAME="ahuber"
declare -x M2="/opt/maven/bin"
declare -x M2_HOME="/opt/maven"
declare -x OLDPWD="/Users/ahuber/Documents/Projects/crazyflie/crazyflie-firmware"
declare -x PATH="/Users/ahuber/Downloads/GCC-ARM/gcc-arm-none-eabi-4_9-2014q4/bin:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin"
declare -x PREV_PATH="/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin"
declare -x PWD="/Users/ahuber/Documents/Projects/crazyflie/crazyflie-firmware"
declare -x SHELL="/bin/bash"
declare -x SHLVL="1"
declare -x SSH_AUTH_SOCK="/private/tmp/com.apple.launchd.wXWlNMtRcy/Listeners"
declare -x SVN_HOME="/Applications/Xcode.app/Contents/Developer/usr/subversion-1.6/bin"
declare -x TERM="xterm-256color"
declare -x TERM_PROGRAM="Apple_Terminal"
declare -x TERM_PROGRAM_VERSION="361.1"
declare -x TERM_SESSION_ID="1B4FB4E1-3C2C-4F7F-9330-9C5FF929F420"
declare -x TMPDIR="/var/folders/rm/yzfbkl4j7k19jxgzrxb27dbw0000gn/T/"
declare -x USER="ahuber"
declare -x XPC_FLAGS="0x0"
declare -x XPC_SERVICE_NAME="0"
declare -x __CF_USER_TEXT_ENCODING="0x1F5:0x0:0x3"

Do you get the same values for these variables if you execute "env" or "printenv"? Could there be any scripts in your .profile, .bashrc, or other changing things in your environment before a build? I'm kinda stumped on this one though I'm pretty sure it has to be and environmental difference. The md5 hash I get on my builds is the same I get in the 0.6 VM. We've got to find the difference in your setup.